View profile weekly - Issue #23: Making data science more useful, deploying AI without technical debt...


nibble dispatch

August 21 · Issue #23 · View online

Curated essays about the future of Data Science. Production Data Science and learning resources for continuous learning. Covers Data Science, Data Engineering, MLOps & DataOps. Curated by people at

This week’s top pick is O'Reilly Data Show’s Make Data Science More Useful where Ben Lorika speaks with Cassie Kozyrkov, Chief Decision Scientist at Google Cloud.

Let's "Make Data Science more useful"
Let's "Make Data Science more useful"
A great model is not enough
Deploying AI without technical debt
An overview of good practices from agile development, DevOps and statistical process control to minimize technical debt, reduces cycle time and improves code and data quality.
AI and Data Science are all the rage, but there is a problem that no one talks about. […] Deploying and maintaining [AI systems] over time is getting exponentially more complex and expensive.
Models for integrating data science teams within organizations
A comparative analysis
This post by Pardis Noorzad, Data Science Manager at Twitter, compares some of the popular models for integrating data science teams within organizations.
Designing and building a data science team is a complex problem; so is determining the nature of interactions between data scientists and the rest of the organization.
How bias distorts AI
The silent killer of AI
An interview with Dr. Rebecca Parsons, CTO at ThoughtWorks, on the consequences of biased AI and how to deal with it.
If we allow this incredible technology to continue to advance but fail to address questions around biases, our society will undoubtedly face a variety of serious moral, legal, practical and social consequences. It’s important we act now to mitigate the spread of biased or inaccurate technologies.
Trends in Natural Language Processing
ACL 2019 In Review
ACL is the Annual Meeting of the Association for Computational Linguistics. You probably won’t have the time to review all the work presented at the conference, but this overview by Mihail Eric, Machine Learning scientist at Alexa AI, is a very good start.
PyTorch 1.2.0 released
Contains a significant amount of effort in areas spanning JIT, ONNX, Distributed, as well as Performance and Eager Frontend Improvements.
On a side note, Facebook also launched an online Global Pytorch Hackathon with $61,000 in prizes.
How to become more marketable as a data scientist
For this analysis, the team from cvcompiler looked at 300 Data Science vacancies from StackOverflow, AngelList, and similar websites.
How to manage impostor syndrome in data science
The best way to overcome it is to understand it
Impostor syndrome is the elephant in the data science lab. Everyone has it, no one thinks other people have it, and no one talks about it.
Top 10 blog posts to help you transition to data engineering
A good list covering getting started, projects and framemork specifics articles.
The industry demand for Data Engineers is constantly on the rise, and with it more and more software engineers and recent graduates try to enter the field. The biggest hurdle for newcomers lies in understanding the Data Engineering landscape and getting hands-on experience with relevant frameworks.
Learning Resources
Parameter optimization in neural networks
This interactive visualizations will help you develop an intuition for setting up and solving the optimization of a model parameters.
Dask tutorial
Dask is very exciting and this tutorial is one of the best to get started with it and understand both high-level and low-level use of Dask. This material was used in this video from SciPy 2018.
Did you enjoy this issue?
If you don't want these updates anymore, please unsubscribe here.
If you were forwarded this newsletter and you like it, you can subscribe here.
Powered by Revue